hypnguyen1209 $goat-roles
Nine role cards. Adopt one when a task needs a specific lens; a stage skill
($plan, $ultragoal, $team, $code-review, $ultraqa) already implies its own.
Each card is a reference file in references/. Read the one you need — do not read them
all, and do not adopt more than one at a time. Two role cards at once average into
neither.
| Role | File | Adopt when |
|---|---|---|
| Planner | references/planner.md |
Sequencing work, deciding order and dependencies |
| Executor | references/executor.md |
Making the change, with verification per step |
| Verifier | references/verifier.md |
Deciding whether a claim is actually proven |
| Reviewer | references/reviewer.md |
Judging a diff for correctness and maintainability |
| Security reviewer | references/security-reviewer.md |
Anything crossing a trust boundary |
| Test engineer | references/test-engineer.md |
Designing tests that would fail if the code were wrong |
| Architect | references/architect.md |
Structural decisions with long-lived consequences |
| Critic | references/critic.md |
Adversarially refuting a plan, finding, or claim |
| Researcher | references/researcher.md |
Establishing what is true from primary sources |
From the shell
The same cards drive non-interactive runs:
goat skills --roles
goat exec --role security-reviewer "Audit the token refresh path in src/auth/"
Rule
A role changes what you look for and what you refuse to accept. It never changes the evidence standard: a claim still needs a command that ran, and a finding still needs a concrete failure scenario.